INDEX

Explanations

references to scientific testing and results

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

linkplain

-0.16

aler

-0.15

getti

-0.15

uits

-0.14

 odds

-0.14

ÐµÑĢÐ¿

-0.14

 inout

-0.14

 link

-0.14

($

-0.14

ija

-0.14

POSITIVE LOGITS

 OECD

0.20

 dose

0.17

nom

0.16

 blank

0.16

 test

0.16

å¶

0.16

 ÙħØ¨

0.16

akk

0.15

uplicates

0.15

ENU

0.15

Activations Density 0.003%

references to scientific testing and results

No Comments

No Known Activations